-
Bug
-
Resolution: Done
-
Normal
-
ACM 2.8.3
-
1
-
False
-
None
-
False
-
-
-
MCO Sprint 27, MCO Sprint 28, MCO Sprint 29
-
-
-
None
Description of problem:
In the scenario where a managed cluster needs to be manually cleaned due to a downed hub (https://docs.redhat.com/en/documentation/red_hat_advanced_cluster_management_for_kubernetes/2.10/html/clusters/cluster_mce_overview#removing-a-cluster-from-management-in-special-cases) we need instructions for how to properly clean the ACM Observability component as well.
If we simply remove all ACM namespaces and CRs including open-cluster-management-addon-observability then this does not account for the modifications made by ACM Observability to the cluster-monitoring-config. This results in the cluster-monitoring-config retaining its previous configuration to send metrics to the old hub and is not overwritten by the manifestwork for the new hub.
Version-Release number of selected component (if applicable):
2.8
How reproducible:
Reproducible
Steps to Reproduce:
With ACM1 as the Active Hub and ACM2 as the Passive Hub when ACM1 is taken down we then remove all ACM projects and CRDs as instructed by the above documentation from the managed cluster. We then attempt to attach the managed cluster to ACM2 which accepts the import without issue but we later see alerts trying to be sent back to ACM1 due to the cluster-monitoring-config retaining its old settings.
Actual results:
Expected results:
Additional info:
- links to
-
RHSA-2024:135868 Red Hat Advanced Cluster Management 2.12.0 security and bug fixes